Patient-reported outcomes (PROs) have gained increasing recognition as legitimate biomedical endpoints and in 2004 NIH launched the PROMIS Initiative to improve outcome measurement by building and validating item banks. If the promise of using patient-reported outcomes to monitor and improve health is to be realized, it is essential to ensure that PROMIS tools are appropriate and psychometrically equivalent for diverse populations cared for outside of specialized health care settings. This is especially important given the substantial health disparities across race/ethnic, age and social class groups. However, there has been limited research establishing equivalence of measures across diverse groups. To fill this gap, we propose a 3-pronged approach to evaluate PROMIS item bank measures in a population-based sample of White, Hispanic, Black and Asian adult cancer patients. We focus on cancer since it is a leading chronic disease and one where there are important disparities in experience and outcomes across the disease trajectory. We have selected 5 cancers (female breast, uterine and cervical cancer, and male and female colorectal cancer and non- Hodgkin's Lymphoma) to ensure a wide age range and symptoms. We use IRT-based methods and construct validation to assess the use of item bank measures for the domains of pain, depression, fatigue, social function, and physical function. In our first phase, our experienced, multidisciplinary team will conduct a qualitative study to assess conceptual equivalence of PROMIS item banks, focusing primarily on Latino and Asian cancer patients since there has been little research in these groups; we will also translate items into Chinese. In our next phase, we will conduct a population based study using 4 cancer registries (from California (2), New Jersey, Louisiana) to randomly recruit 4,000 patients diagnosed in the previous 6-9 months. We will interview 1,000 patients from each race/ethnic group divided evenly among two age groups (ages 21-64, and ages 65-84). Participants will be surveyed using a mixed mode approach (mail and phone) at initial recruitment and again 6 months later. Our third component is an exploratory clinical study (n=200) to examine whether the PROMIS fatigue measures generate sufficiently precise estimates of change over time during the first 6 months of chemotherapy to help guide care management. Our study will provide a comprehensive set of empirical data on PROMIS measurement performance across sub- populations, indicating likely causes of any observed differences. Our approach leverages investments in, and extends the scope of, the PROMIS Initiative by developing measures that can be used to evaluate the impact of intentions, inform practice, and monitor population health disparities.